How to extract base64 code out of a file?

[postcd]

Dear members im having text file which contains 1 or 2 base64 blocks of code. Following is part of the file and 1 base64 block.

Content-Transfer-Encoding: base64

RGVhciBMdWlzLA0KICANCkdvb2QgbW9uaW5nIQ0KT3VyIEJhbmsgYWNjb3VudCBpcyB1bmRlcmdvaW5nIGF1ZGl0IGZvciB0aGUgZmlzY2FsIHllYXIgYXMgcmVxdWlyZWQgYnkgdGhlIGNoaW5hIGZvcmVpZ24gY3VycmVuY3kgY29udHJvbCBwb2xpY3kgYW5kIHRoZSBhY2NvdW50IHdpbGwgbm90IGJlIGF2YWlsYWJsZSB1bnRpbCB0aGUgQ2hpbmVzZSB0YXggYnVyZWF1cyBhcmUgc
2F0aXNmaWVkIHRoYXQgYWxsIGFwcGxpY2FibGUgdGF4ZXMgaGF2ZSBiZWVuIHBhaWQgdXAuIFBsZWFzZSBob2xkIHBheW1lbnQgc28gaSBjYW4gZnVybmlzaCB5b3Ugd2l0aCBvdXIgcmV2aXNlZCBiYW5raW5nIGluZm9ybWF0aW9uLiAgDQoNCiAgDQpCZXN0IHJlZ2FyZHMNCiANCkRhdmlk

------=_Part_143209_644876817.1451544132767--

Being on Linux, which command i can do to extract first base64 out of the file? My aim is to decode output to be readable

---------

Example file decode command: http://superuser.com/a/663397
Example phrasse decode command: http://askubuntu.com/a/178546

but i do not know how to extract only the base64 phrasse out of file which also contains non base64 contents.

 

[unixfish]

I would normally do this in Perl using the MIME::Base64 module.

But you asked for a Linux command, and after a quick Google search, I found:

echo [encodedstring] | base64 --decode

from coreutils, or

openssl enc -base64 -d <<< [encodedstring]

using openssl (you will need the -A option if the input is more than 64 characters).

Either way, you will need to separate the MIME string from the rest of the text file to translate it.

 

[postcd]

you will need to separate the MIME string from the rest of the text file to translate it.

Yes, decoding is not the problem i also found commands, but that separation of non base64 and base64 thats what i dont know how to do ... as mentioned in my first post

 

[unixfish]

Ah - you did say that. I was not paying close enough attention to your post - my fault!

From a command line, my thoughts seems convoluted - I tend to go to Perl for issues like this. It seems from your example you have a header and a trailer line, which makes the beginning and end of the MIME code easy to find programatically.

From a shell, you could use "wc -l" on the file to get the total number of lines, use

start=`awk '/header-text/{ print NR; exit }' input-file`

to get the line number where the text starts (or add 1 or more to the actual text), then do the same for the end line, and finally a combination of head and tail with some calculations of lines stringing them together, and pipe this into the decoder command.

This seems like a lot of work to me - there are ways of doing this with awk, but you are probably going to write an awk script to pass the file through.

I have done a fair amount of MIME encoding / decoding in Perl, so it just seems easier to me to write a small program to accept the file name as a parameter and print out the decoded MIME strings.

$0.02. Take this for what it's worth - my opinion.